📄 Technical report | 💻 Cookbooks | 👀 Atla agent evals

@keyframes rainbow { 0% { background-position: 0% 50%; } 50% { background-position: 100% 50%; } 100% { background-position: 0% 50%; } }

Model Summary

Atla Selene Mini is a state-of-the-art small language model-as-a-judge (SLMJ). Selene Mini achieves comparable performance to models 10x its size, outperforming GPT-4o on RewardBench, EvalBiasBench, and AutoJ.

Post-trained from Llama-3.1-8B across a wide range of evaluation tasks and scoring criteria, Selene Mini outperforms prior small models overall across 11 benchmarks covering three different types of tasks:

• Absolute scoring, e.g. "Evaluate the harmlessness of this response on a scale of 1-5"
• Classification, e.g. "Does this response address the user query? Answer Yes or No."
• Pairwise preference. e.g. "Which of the following responses is more logically consistent - A or B?"

It is also the #1 8B generative model on RewardBench.

The 70B version of this model, Selene 1, is on Hugging Face. Get started with the world's most powerful evaluation model here.

Model Details

• Developed by: Atla
• Model type: Post-trained from Llama-3.1-8B
• Language(s) (NLP): Primarily English but supports German, French, Italian, Portuguese, Hindi, Spanish, Thai
• Context length 128K

Model Use

Selene Mini can be used as a general-purpose evaluation model. It supports different inputs & scoring scales, generates structured evaluation outputs, and provides qualitative critiques with reasoning.

Try our cookbooks to get started with two popular use cases below:

• Absolute scoring
• RAG hallucination

To achieve best results, we provide the prompts we used for training here.

Remember to apply the conversation template of Llama 3 - not doing so might lead to unexpected behaviors. You can find the conversation class at this link or you can refer to the below code that will apply it.

Quickstart (HF Transformers):

from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto

model_id = "AtlaAI/Selene-1-Mini-Llama-3.1-8B"

model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_id)

prompt = "I heard you can evaluate my responses?" # replace with your prompt / we provide prompt templates used during training at github.com/atla-ai/selene-mini/tree/main/prompt-templates
messages = [{"role": "user", "content": prompt}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
model_inputs = tokenizer([text], return_tensors="pt").to(device)

generated_ids = model.generate(model_inputs.input_ids, max_new_tokens=512, do_sample=True)
generated_ids = [output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Contact

support@atla-ai.com
You can also join our Discord!

Citation

If you are using the model, cite using

@misc{alexandru2025atlaseleneminigeneral,
      title={Atla Selene Mini: A General Purpose Evaluation Model}, 
      author={Andrei Alexandru and Antonia Calvi and Henry Broomfield and Jackson Golden and Kyle Dai and Mathias Leys and Maurice Burger and Max Bartolo and Roman Engeler and Sashank Pisupati and Toby Drane and Young Sun Park},
      year={2025},
      eprint={2501.17195},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2501.17195}, 
}